We propose to localize the neuroanatomical source generators of the magnetic evoked field (EF) components M30, M50, M100, M300, which we believe are the magnetic correlates of auditory evoked potential (EP) components P30, P50, N100, and P300, several of which have been found to be abnormal in schizophrenic patients. We will perform the localizations by mapping extra-cranial magnetic fields utilizing a multi-channel SQUID gradiometer system, mathematically estimating intracranial source locations, and identifying the structures extent at those locations utilizing magnetic resonance (MR) imaging. Studies will be performed in normals, schizophrenic patients and subjects at genetic risk for schizophrenia. We will further determine if abnormalities previously described in EP components are also found with EF components. We will perform MEG EF recordings in patients with MR-defined lesions in temporal lobe auditory and limbic structures to determine if defined lesions impact of EF generation in the predicted manner. We will also map extra-cranial MEG EFs simultaneously with EPs recorded from stereotaxically implanted depth electrodes in alert M. nemestrina monkeys to correlate the depth EP potential distribution with surface recorded MEG field topographer. Estimated source locations in the monkeys will be reversibly lesioned, and ultimately histological examinations will be performed to correlate depth EP, surface EF, lesion data, and MR data. We will also perform a series of modeling studies utilizing simple and complex head and source geometries to improve techniques for localization of subcortical current sources utilizing EF mapping. Our findings will have direct relevance to a better understanding of temporal lobe abnormalities described in schizophrenic patients, and will permit the correlation of EP and EF abnormalities with dysfunction at specific neuroanatomical sites. We believe our results may be useful in the diagnosis of schizophrenia and the prediction of the probability of developing the disorder in those genetically at risk.